HS1-associated protein X-1 interacts with membrane-bound IgE: impact on receptor-mediated internalization.

Abstract

Engagement of the BCR triggers signals that control affinity maturation, memory induction, differentiation, and various other physiological processes in B cells. In previous work, we showed that truncation of the cytoplasmic tail of membrane-bound Ig (mIg)E in vivo resulted in lower serum IgE levels, decreased numbers of IgE-secreting plasma cells, and the abrogation of specific secondary responses correlating with a defect in the selection of high-affinity Abs during the germinal center reaction. We concluded that the Ag receptor is necessary at all times during Ab responses not only for the maturation process, but also for the expansion of Ag-specific B cells. Based on these results, we asked whether the cytoplasmic tail of mIgE, or specific proteins binding the cytoplasmic tail in vivo commit a signal transduction accompanying the B cell along its differentiation process. In this study, we present the identification of HS1-associated protein X-1 as a novel protein interacting with the cytoplasmic tail of mIgE. ELISA, surface plasmon resonance analysis, and coimmunoprecipitation experiments confirmed the specific interaction in vitro. In functional assays, we clearly showed that HS1-associated protein X-1 expression levels influence the efficiency of BCR-mediated Ag internalization.

Full text

Page 1

HS1-Associated Protein X-1 Interacts with Membrane-Bound
IgE: Impact on Receptor-Mediated Internalization1
Iris Oberndorfer,* Doris Schmid,* Roland Geisberger,* Gertrude Achatz-Straussberger,*
Reto Crameri,†Marinus Lamers,‡and Gernot Achatz2*
Engagement of the BCR triggers signals that control affinity maturation, memory induction, differentiation, and various other
physiological processes in B cells. In previous work, we showed that truncation of the cytoplasmic tail of membrane-bound Ig
(mIg)E in vivo resulted in lower serum IgE levels, decreased numbers of IgE-secreting plasma cells, and the abrogation of specific
secondary responses correlating with a defect in the selection of high-affinity Abs during the germinal center reaction. We
concluded that the Ag receptor is necessary at all times during Ab responses not only for the maturation process, but also for the
expansion of Ag-specific B cells. Based on these results, we asked whether the cytoplasmic tail of mIgE, or specific proteins binding
the cytoplasmic tail in vivo commit a signal transduction accompanying the B cell along its differentiation process. In this study,
we present the identification of HS1-associated protein X-1 as a novel protein interacting with the cytoplasmic tail of mIgE. ELISA,
surface plasmon resonance analysis, and coimmunoprecipitation experiments confirmed the specific interaction in vitro. In func-
tional assays, we clearly showed that HS1-associated protein X-1 expression levels influence the efficiency of BCR-mediated Ag
internalization. The Journal of Immunology, 2006, 177: 1139–1145.
F
the BCR consists of the membrane-bound Ig (mIg),3a tetramer of
a H chain homodimer and either two ? or two ? L chains. Two
further domains, the membrane-spanning domain (coded by exon
M1) and the cytoplasmic tail (coded by exon M2), provide the
prerequisite for later membrane localization. All mIg isotypes are
noncovalently associated with a heterodimer of two transmem-
brane proteins, Ig? (CD79a) and Ig? (CD79b) (1). The Ig?/Ig?
molecules belong to the Ig superfamily and carry ITAM (consen-
sus: YxxL/Ix6–8YxxL/I) in their cytoplasmic tails (2, 3). Receptor
stimulation initiates a complex cascade of cytoplasmic-signaling
events resulting in the appropriate cellular response. Interestingly,
none of the cytoplasmic tails of the various Ig isotypes contain
ITAM motifs or ITIMs (consensus: I/V/L/SxYxxL/V) (4). Thus, a
signaling cascade committed by the cytoplasmic tails of mIgs itself
seems to be unlikely, but rather postulates proteins that actively
bind the cytoplasmic tails of mIgs (5–7).
rom the numerous surface markers of a B cell, the BCR is
probably the most powerful one in influencing develop-
mental processes. The extracellular Ag-binding moiety of
The earliest detectable biochemical event that follows BCR ag-
gregation is increased activity of protein tyrosine kinases of the Src
family (Lyn), resulting in phosphorylation of the tyrosine residues
within the ITAMs of Ig? and Ig? and the non-ITAM tyrosine 204
of Ig? (8), subsequently followed by the initiation of several sig-
naling pathways (reviewed in Refs. 6, 9, and 10). Syk, after acti-
vation by Lyn, tyrosine-phosphorylates the hemopoietic linage
cell-specific protein 1 (HS1), which is specifically expressed in
hemopoietic cells. HS1 is a 75-kDa hemopoietic linage-specific
protein (11, 12), with known functions in B cell proliferation and
BCR-induced apoptosis (11). After tyrosine phosphorylation, HS1
translocates from the cytoplasm to the nucleus (11). HS1 was
shown to interact with HS1-associated protein X-1 (HAX-1, also
known as HS1-binding protein) (13), a 35-kDa, ubiquitously ex-
pressed protein. However, the subcellular localization of HAX-1
depends on the cell type. Accordingly, HAX-1 was localized in the
cytoplasm (13) and even in mitochondria (14), and was found
along the endoplasmic reticulum and the nuclear envelope as well
(15). HAX-1 displays a weak homology to the intracellular mam-
malian protein, BNip3 (13), and similarity to the functionally im-
portant domains BH1 and BH2 of Bcl-2 family member proteins.
An internal proline-glutamic acid-serine-threonine sequence sug-
gests that the HAX-1 protein may be degraded rapidly (16).
Besides its association with HS1, HAX-1 was also reported to
physically associate with several other molecules indicating that the
biological function of HAX-1 can roughly be divided into three cat-
egories: 1) Association of HAX-1 with HS1 (13), Kaposi’s sarcoma-
associated herpes virus K15 (17), Epstein-Barr nuclear Ag 5 (EBNA5
or EBNA-LP) (18, 19), Bcl-2 and BHRF1 (EBV homolog of Bcl-2)
(20), Omi/HtrA2, a nuclear-encoded mitochondrial serine protease
with proapoptotic function (14) and HIV protein Vpr (21) indicates
the involvement of HAX-1 in the regulation of apoptotic processes. 2)
Association of HAX-1 with cortactin (or EMS1) (15), polycystic kid-
ney disease protein 2 (15), ATP-binding cassette transporters BSEP,
MDR1, and MDR2 (22) and G?-13, a cell migration-stimulating
component (23) emphasizes the involvement of HAX-1 in cell mo-
tility processes.
*Department of Molecular Biology, University of Salzburg, Salzburg, Austria;†Swiss
Institute of Allergy and Asthma Research, Davos, Switzerland; and‡Max Planck
Institute for Immunobiology, Freiburg, Germany
Received for publication March 1, 2006. Accepted for publication April 26, 2006.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1Experimental work was supported by the Austria Science Foundation Projects
P-19017 and T166 (Hertha Firnberg) and the Austrian National Bank (OENB Grant
11710), the Swiss National Science Foundation (Grants 31-63382.00/2 and 310000-
112540), and the OPO Foundation, Zu ¨rich.
2Address correspondence and reprint request to Dr. Gernot Achatz, Department of
Molecular Biology, University of Salzburg, Hellbrunnerstrasse 34, A-5020 Salzburg,
Austria. E-mail address: gernot.achatz@sbg.ac.at
3Abbreviations used in this paper: mIg, membrane-bound Ig; HAX-1, HS-1-associ-
ated protein X-1; EBNA, Epstein-Barr nuclear Ag; DHFR, dihydrofolate reductase;
CH, constant region H chain; Arp, actin-related protein; NP, nitrophenyl; siRNA,
small-interfering RNA; His, histidine; fwd, forward; rev, reverse; SPRA, surface plas-
mon resonance analysis.
The Journal of Immunology
Copyright © 2006 by The American Association of Immunologists, Inc.0022-1767/06/$02.00

Page 2

3. Finally, physical interaction of HAX-1 with the 3? untrans-
lated region of human vimentin mRNA (24) indicates RNA-bind-
ing capacity and, as interaction factor of the IL-1? N terminus
(25), HAX-1 seems to act as cytoplasmic retention factor.
In summary, HAX-1 performs a multifunctional impact on bio-
logical processes. In the current work, we present HAX-1 as novel
interactionpartnerofthecytoplasmictailofmIgE.Weshowevidence
that the cytoplasmic tail of IgE serves as docking site for interacting
proteins thereby affecting Ag processing and presentation.
Materials and Methods
Reagents and Abs
Monoclonal mouse-IgG1 anti-HAX-1 Ab was purchased from BD Trans-
duction Laboratories (BD Biosciences), monoclonal mouse-IgG1 anti-
penta-His Ab from Qiagen, monoclonal mouse IgG1 anti-?-tubulin Ab
from Abcam, and bovine anti-mouse IgG1-conjugated with HRP as well as
rat-anti-CD19-RPE was purchased from Serotec. Polyclonal anti-murine
IgE serum was raised in rabbits against purified mouse IgE, ? (BD Pharm-
ingen) and goat-anti-rabbit-IgG (whole molecule)-alkaline phosphatase
conjugate was purchased from Sigma-Aldrich. Rat (IgG1)-anti-murine IgE
clone 51.3 (anti-CH4-domain of IgE) was provided by Dr. T. Waks (De-
partment of Immunology, The Weizmann Institute of Science, Rehovot,
Israel) (26).
Cell lines and media
J558L-mb1 (gift from Dr. J. Wienands, Department of Biochemistry, Uni-
versity of Bielefeld, Germany.) and JW813/4 BCR?were grown in RPMI
1640 medium supplemented with 10% FCS, 55 ?M 2-ME, 100 U/ml pen-
icillin, and 100 ?g/ml streptomycin. JW813/4 BCR?medium was addi-
tionally supplemented with 25 mg/L mycophenolic acid, 250 mg/L xan-
thine, and 13.6 mg/L hypoxanthine.
Cell line JW813/4 BCR?small-interfering RNA (siRNA) was stably
transfected with pU6-siRNA plasmids (Biomyx Technology) and selected
with 400 ng/ml geneticin (G418 sulfate).
Production and purification of recombinant proteins
DHFR, CH4, DHFR tail, CH4 tail. The cytoplasmic tail of mIgE (YG
ATVTVLKVKWVFSTPMQDTPQTFQDYANILQTRA)
nally fused to the constant region H chain (CH)4 domain of mouse or
mouse dihydrofolate reductase (DHFR), respectively. CH4 and DHFR
functioned as carrier proteins for the tail peptide. The cDNA sequences
were cloned into pHIS-parallel vectors (27) leading to N-terminally His-
tagged proteins. Recombinant His-tagged proteins, designated as CH4,
DHFR, CH4 tail, and DHFR tail, were batch-purified under denaturing
conditions in the presence of 8 M urea using nickel-nitrilotriacetic acid
agarose.
Murine HAX-1. Murine HAX-1 cDNA-sequence was amplified from a
murine cDNA-library by PCR using primers: no. 288 forward (fwd), 5?-
ATGAGCGTCTTTGATCTTTTCCGAGGC and no 289 reverse (rev),
5?-CTATCGGGACCGAAACCAACGTCCTAG.
The PCR product was ligated into the pPCR-Script AMP SK?plasmid
(Stratagene) according to the manufacturer’s protocol. After sequence con-
firmation HAX-1 was recloned into the pHIS-parallel 2 vector. After nick-
el-nitrilotriacetic acid purification, the protein was dialyzed against 10 mM
NaH2PO4(pH 7.5).
Construction of a murine B cell cDNA library in a pJuFo-
phagemid vector
wasC-termi-
Female BALB/c mice were repeatedly immunized with 5 ?g of recombi-
nant Bet v 1a in 3–4 wk intervals. Lymph nodes (axial and inguinal) and
the spleen were prepared from two mice erythrocyte lysis was performed.
The splenocytes and the monodisperse lymph node cell preparation were
pooled and cultured at a density of 106cells/ml in RPMI 1640 complete me-
dium supplemented with 15 ?g/ml LPS, 10 ?g/ml recombinant Bet v 1a, and
500 U/ml recombinant mouse IL4 (BD Pharmingen). After 3 days, lympho-
cytes were stained with anti-CD19-PE Ab and CD-19?B cells were sorted
with MACS technique yielding ?4 ? 107CD-19?B cells of ?95% purity.
mRNA was isolated with MACS oligo(dT) Micro beads. Isolated mRNA was
transcribed into cDNA using the Stratagene cDNA synthesis kit. The cDNA
was ligated (EcoRI/XhoI) into a modified pJuFo vector (28), designated as
pGA110 (5). The library contained ?5 ? 105single transformants with insert
size ranging between 1400 and 2400 bp. Inserts were sequenced using the
following primers: no. 220 fwd, 5?-GCAAACCGAAATCGCGAACCTGC
and no. 214 rev, 5?-GTAAAACGACGGCCAGTG.
Phage display biopanning
Phage was generated as described previously (28, 29, 7). The cytoplasmic
tail was fused to the carrier proteins DHFR or CH4 domain of IgE, re-
spectively. Carrier-specific phage was depleted by preincubation of the
phage preparation with an excessive amount of the respective carrier pro-
tein before each panning round. Four different biopanning experiments
with different bait proteins were performed. Decreasing amounts of recom-
binant tail fusion proteins (200, 150, 100, and 100 ?g in the first, second,
third, and fourth panning round, respectively) were immobilized on Maxi-
Sorp Immuno tubes. Immobilization was done in denaturing buffer (8 M
urea, 50 mM NaH2PO4(pH 8), 300 mM NaCl) over night at 4°C. Proteins
were refolded “in situ” by washing the immunotubes once with 5 M urea
in PBS (pH 7.5) followed by washing twice with PBS. For the phage
selection, preincubated phage was transferred into the blocked tubes.
Eluted phage was used for infection of TG1 Escherichia coli cells for a
successive panning round. Specific phage were detected with a “Phage”
ELISA by adding serially diluted phage (3.511to 1.110PFU/ml) to the CH4
tail or CH4-coated wells and subsequent staining with anti-M13-HRP Ab
(Amersham Pharmacia Biotech). OD values measured with CH4 as Ag
were subtracted from signals measured with the CH4 tail to obtain tail-
specific signals.
Real-time PCR
Measurement of HAX-1 enrichment in phage display. The relative
amounts of HAX-1-encoding phagemids obtained after each panning round
were determined by real-time PCR using a HAX-1-specific primer-set: no.
292 fwd, 5?-TAGACAGTGAGGGCCGGAGGGAGAC; no. 293 rev, 5?-
TGGCAATGGGCAACAGGAAGGGAGTGG; and a pGA110 vector-spe-
cific primer set: no. 290 fwd, 5?-CGGCGGCTCTGGTGGTGGTTCT; no.
291 rev, 5?-ACTGTAGCGCGTTTTCATCGGCATTTTCGGTCAT.
Phagemid preparations obtained from bacteria infected with phage from
each panning round were used as templates for PCR amplification. To
prove correct amplification, PCR products were sequenced. Values ob-
tained with the vector-specific primer set were set to 100% and the per-
centage of HAX-1-encoding plasmids was calculated for each sample. All
individual PCRs were done in duplicates and SDs were calculated from
three independently performed experiments.
Measurement of HAX-1 transcripts in siRNA-transfected cells lines.
The relative amounts of HAX-1 mRNA in cells transfected with siRNA
targets for HAX-1 were determined using primer set (nos. 292, 293) for
HAX-1 and an Arbp (60S acidic ribosomal protein P0)-specific primer set:
fwd, 5?-TGCACTCTCGCTTTCTGGAGGGTG; rev, 5?-AATGCAGAT
GGATCAGCCAGGAAGG.
mRNA was isolated from 107cells, cDNA was prepared from 3 ?g of
mRNA and finally used as template in the real-time PCR amplifications.
Coimmunoprecipitation
“In vitro” coimmunoprecipitation. A total of 2 ? 107J558L cells were
resuspended in 1 ml of ice-cold lysis buffer (25 mM Tris (pH 7.5), 140 mM
NaCl, 1 mM EDTA, 0.5% Nonidet P-40, 1 mM sodium orthovanadate)
containing a protease inhibitor mix (Complete Mini; Roche). A total of 30
?g of carrier-tail-fusion protein (CH4 tail) or carrier protein (CH4), re-
spectively, were added and cells were then lysed by rotating at 4°C for 45
min. Insoluble material was removed by centrifugation at 15,000 ? g for
10 min at 4°C. A total of 2 ?g of anti-5xHis Ab were added to the cleared
lysate and left rotating at 4°C for 2 h. For precipitation 35 ?l of protein
G-Sepharose 4 Fast Flow (Amersham Biosciences) were added and the
mixture left rotating for 2 further hours at 4°C. Sepharose was pelletized by
centrifugation at 250 ? g at 4°C for 5 min, and the precipitate washed
twice with 1 ml of ice-cold lysis buffer and twice with 1 ml of ice-cold
wash buffer (100 mM Tris (pH 8.1), 0.5 M LiCl). Immune complexes were
eluted from the beads by incubation with 50 ?l of 2? reducing SDS-
PAGE-loading buffer (125 mM Tris, 140 mM SDS, 20% glycerol, 10%
2-ME, 30 ?M bromphenol blue (pH 6.75)) for 5 min at 95°C. After re-
moving beads by centrifugation at 15,000 ? g for 3 min at 4°C, 20 ?l of
the cleared supernatant were loaded onto a 10%-SDS-PAGE.
“In vivo” coimmunoprecipitation. A total of 2 ? 107JW813/4 BCR?
and JW813/4 BCR?myeloma cells were lysed and the precipitation was
performed as described before. Nitrophenyl (NP)-specific IgE-BCR was
precipitated using 30 ?l of NP-Sepharose (4-hydroxy-3-nitrophenylacetyl
hapten conjugated to Sepharose via a six-carbon spacer; Biosearch Tech-
nologies) or CL-2B Sepharose (Stratagene) as a control for unspecific bind-
ing to Sepharose.
Crude cell lysates were separated by SDS-PAGE and transferred to an
Immobilon-P polyvinylidene difluoride (0.45 ?m) membrane (standard
1140 mIgE-INTERACTING PROTEINS

Page 3

Western protocol). For detection, primary and HRP-conjugated secondary
Abs, see Reagents and Abs, were used.
Surface plasmon resonance analysis (Biacore X)
Recombinant His-tagged murine HAX-1 (60 ng/?l in 10 mM NaH2PO4
(pH 7.5)) was diluted 1/5 in 10 mM sodium acetate (pH 4) and coupled to
the CM5 chip according to the manufacturer’s instructions. Approximately
4000 resonance units rHAX-1 were coupled to flow cell 2. Empty flow cell
1 served as reference. The synthesized IgE-tail peptide (1 mg/ml in PBS;
sequence HHHHHHKVKWVFSTPMQDTPQTFQDYANILQTRA) was
injected at different concentrations (5–100 ?M) in 10 mM HEPES (pH
7.4), 150 mM NaCl, 3 mM EDTA, 0.005% surfactant P20 buffer (Biacore),
and the surface plasmon resonance was recorded. Data were analyzed with
BIAevaluation software (Biacore).
RNA interference
Two siRNA target sequences were selected via ?rnaidesigner.invitrogen.
com? (target 1) and ?www.ambion.com/techlib? (target 8). The control tar-
get sequence was taken from Ortiz et al. (22). Selected target sequences
were tested for exact matches of ?12 nucleotides. Target sequences that
matched mRNA sequences different from the murine HAX-1 sequence
were excluded. The siRNA target sequence was cloned in forward and
reverse complement orientation spaced by a loop with the sequence
TTCAAGAGA. The expression of this construct leads to a RNA stem loop
conformation, which contains the desired RNA double strand needed for
RNA interference. For cloning purposes, restriction sites at the 5? and 3?
ends were added.
HAX-1 siRNA target 1: nos. 433 and 434. No. 433, 5?-TTTGGCTAC
TAGGACGTTGGTTTCGTTCAAGAGACGAAACCAACGTCCTAGT
AGCTTTTT; no. 434, 5?-CTAGAAAAAGCTACTAGGACGTT
GGTTTCGTCTCTTGAACGAAACCAACGTCCTAGTAGC.
HAX-1-siRNA target 8: nos. 435 and 436. No. 435, 5?-TTTGGCTTA
AGTACCCAGATAGTTTCAAGAGAACTATCTGGGTACTTAAGCTT
TTT; no. 436, 5?-CTAGAAAAAGCTTAAGTACCCAGATAGT
TCTCTTGAAACTATCTGGGTACTTAAGC.
Control target: nos. 347 and 438. No. 347, 5?-TTTGGTGTACAGC
GATGTTGTCGTTCAAGAGACGACAACATCGCTGTACACTTTTT;
no.438,5?-CTAGAAAAAGTGTACAGCGATGTTGTCGTCTC
TTGAACGACAACATCGCTGTACAC.
Oligonucleotides were annealed and ligated into the double-digested
(XbaI/BbsI)pU6-siRNA expression
JW813/4 BCR?myeloma cells were transfected and single transfectants
were isolated via limiting dilution in selective medium with geneticin. Sta-
ble integration was tested by genomic PCR with a primer set (nos. 446 and
447) spanning the sequence from the U6 promoter to the siRNA-target
sequence. No. 446 fwd, 5?-GAAGCATTTATCAGGGTTATTGTCT; no.
447 rev, 5?-TTGAGCGTCGATTTTTGTGATG.
All stable cell lines were controlled for the induction of an IFN response
by measuring the induction of OAS1, a classic IFN target gene, by real-
time PCR (nos. OAS1f and OAS1r). OAS1f fwd, 5?-TCCCAACTC
CCGGGCTCTGAG; OAS1r rev, 5?-GCGGGGTACGCCCACTGATG.
vector (BiomyxTechnology).
BCR-internalization assay
Ag-internalization assays were performed with slight modifications accord-
ing to Aluvihare et al. (30). A total of 1.2 ? 106logarithmically growing
JW813/4 BCR?siRNA cells were washed once with PBS and subse-
quently incubated on ice with anti-IgE-FITC (clone EM95-3) for 45 min in
a total volume of 600 ?l of cold FACS buffer (PBS 3% FCS). The cells
were washed twice with ice-cold FACS buffer, resuspended in 1.2 ml of
RPMI 1640 complete medium (10% FCS) and equal aliquots of 0.3 ml
were incubated at 37°C in a water bath in humidified atmosphere contain-
ing 7% CO2in an open FACS tube for the time indicated (0, 10, 20, or 30
min). The cell samples were pelletized and resuspended in 500 ?l of FACS
buffer each. Fluorescence intensities were analyzed using flow cytometry
(FACSCalibur; BD Biosciences). The cells gated for living cells (propidium
iodide). The relative fluorescence intensities were calculated by equating the
fluorescence (geometrical mean) at time 0 with 100%. For calculating the
amount of internalized Ag, the mean fluorescence of time 0 was set to 0%.
Mean values and SDs were calculated from 10 independent experiments.
Results
Identification of HAX-1 as interaction partner of the cytoplasmic
tail of mIgE
We used the phage display biopanning technique to screen a mu-
rine CD19?cDNA library constructed in the modified pJuFo
phagemid vector pGA110 (5, 28). As bait, we used a chimeric
recombinant fusion protein composed of the cytoplasmic tail of
mIgE fused to the C terminus of DHFR or CH4 as carrier (DHFR
tail and CH4 tail). By fusing the short tail with a carrier, we
avoided sterical interference with the blocking agent and in par-
allel were able to mimic a “dimeric IgE tail” during the panning
procedure, as present under in vivo conditions. We point to this
fact because we failed to isolate tail-specific binding partners in a
previous approach using a yeast two-hybrid screen (in this ap-
proach, the bait protein is obligatorily expressed as monomer; our
unpublished data). Thus, we hypothesized that a potential binding
partner might need the tail in its dimeric configuration. Four in-
dependent phage display experiments were performed. In panning
A and B, DHFR and CH4 tails were used during all four panning
rounds. In panning C and D, the carrier protein was exchanged
after the first two panning rounds. After the fourth panning round
of each phage display setup 30 single clones were tested in ELISA
for binding to the recombinant IgE tail. Subsequently, phagemids
from ELISA-positive clones were prepared and sequenced. Ten of
14 sequenced clones represented the C terminus of the HAX-1
protein. Within these 10 clones eight encoded 213 C-terminal aa of
HAX-1 and two clones originated from a longer cDNA clone that
coded for 231 C-terminal aa (Fig. 1). To determine the increase of
HAX-1-encoding phagemids during the four panning rounds we
performed real-time PCR measurements (Fig. 2) using primer sets
that were specific for the phagemid vector backbone or the HAX-
1-cDNA insert. In the original cDNA library, we measured 0.014%
HAX-1-encoding phagemids. The extent of HAX-1 enrichment
varied significantly within the four phage display setups. Only 2%
of all phagemids of the fourth panning round of panning A coded for
HAX-1while62%codedforHAX-1inpanningB.InpanningsCand
D, we measured an intermediate value of ?33%. The variations in
enrichment efficiency may be explained by the varying efficiencies of
the preincubation steps to eliminate carrier-specific phage.
HAX-1-expressing phage recognize the recombinant mIgE-tail in
ELISA
Phage, displaying either the 213 aa- or the 231 aa-long C-termi-
nally truncated HAX-1 protein (Fig. 2) were tested in ELISA for
in vitro binding of the recombinant mIgE tail. As negative control,
we used the Niemann Pick C2 protein. ELISA plates were coated
with 100 ng of purified recombinant protein per well (CH4 or CH4
tail, respectively). Specific phage were detected with an anti-M13-
HRP Ab. Absorbance values were recorded at 405 nm. OD values
measured with CH4 as Ag were subtracted from signals measured
FIGURE 1.
ing frame. Upper panel, HAX-1 shows similarities to the functionally impor-
tant domains of antiapoptotic proteins BH1 and BH2 (BH. . . Bcl-2 homology
domain). The PEST sequence is a sequence motif for rapidly degraded pro-
teins. Lower panel, Two phage-clone populations encoding 213 or 231 aa of
the C-terminal part of HAX-1, respectively, were enriched during phage dis-
play. The alignment with the full-length protein shows that both clones harbor
BH2 and the longer clone partially includes the BH1 domain.
Functionally important domains inside the HAX-1 open read-
1141 The Journal of Immunology

Page 4

with the CH4 tail to obtain tail-specific signals. Both HAX-1 clones
bound specifically to the tail but phage particles displaying the longer
HAX-1 fragment gave higher signals than the shorter (Fig. 3).
HAX-1 binds the IgE tail with moderate affinity
The association between the synthetically produced histidine
(His)-tagged mIgE-tail and rHAX-1 was further tested by surface
plasmon resonance analysis (SPRA) using the Biacore X device.
Four thousand resonance units of rHis-tagged murine HAX-1 was
coupled to flow cell 2 of a CM5 chip. The empty flow cell 1 served
as reference. The synthesized IgE tail peptide was injected at dif-
ferent concentrations (5–100 ?M) and the surface plasmon reso-
nance was recorded. Data were analyzed with BIAevaluation soft-
ware. We calculated an association constant (KA) of 1.4 ? 104
M?1and a dissociation constant (KD) of 7.4 ? 10?5M from the
respective curves (Fig. 4). The calculated KDof 7.4 ? 10?5M
indicated a biologically relevant association.
Endogenous HAX-1 coprecipitates with the recombinant mIgE
tail in vitro
To investigate whether full-length HAX-1 interacts with the re-
combinant mIgE-tail, we performed coimmunoprecipitation exper-
iments. The recombinant fusion protein of tail and carrier (CH4
tail) was added to the lysate of J558L myeloma cells (in vitro
precipitation). As control, we performed the same precipitation
with the carrier protein (CH4) alone. Anti-5xHis Ab was used to
precipitate the His-tagged recombinant proteins. Endogenous mu-
rine HAX-1 was detectable by Western blot analysis only in the
precipitate with the tail protein (CH4 tail) and not in samples pre-
cipitated with the carrier protein alone (Fig. 5).
HAX-1 coprecipitates with mIgE in the cell line JW813/4 BCR?
To investigate whether full-length HAX-1 interacts with the en-
dogenously expressed mIgE tail, we used the JW813/4 BCR?cell
line. This cell line was stably transfected with a mIgE H chain
construct that associated with the endogenous ? L chain to form an
NP-specific BCR of class IgE on the surface of the plasma mem-
brane (1). This mIgE molecule was precipitated with NP-Sepha-
rose. To exclude any BCR-mediated signaling and thus any de
novo phosphorylation events, the specific Ag (NP-Sepharose) was
FIGURE 4.
mIgE tail in SPRA. The rHAX-1 protein was immobilized on a CM5 chip,
and the mIgE-tail peptide (concentrations from 5 to 100 ?M) was used as
soluble analyte. The KA? 1.4 ? 1041/M and KD? 7.4 ? 10?5M were
calculated from the association and dissociation curves using the Langmuir
binding model.
Association and dissociation curves between HAX-1 and
FIGURE 2.
PhageelutedaftereachpanningroundwereusedtoreinfectE.coli.Phagemids
werepreparedfromthedifferentpopulationsandusedastemplateforreal-time
PCR. HAX-1 was detected with primers specific for HAX-1 cDNA, whereas
the plasmid backbone was detected with primers specific for a vector-encoded
phage envelope protein (pIII). The relative percentage of HAX-1 in each sam-
ple was calculated. The measurement of each sample was performed in du-
plicates, and the real-time PCR was performed independently three times.
Panning round “0” corresponds to the original cDNA library.
Enrichment of HAX-1-encoding phage during phage display.
FIGURE 3.
performed a “phage- ELISA” to confirm the interaction between the re-
combinantly expressed mIgE tail and the selected HAX-1-expressing
phage. Phage, expressing the longer HAX-1-cDNA clone (231 C-terminal
aa) or the shorter HAX-1-cDNA clone (213 C-terminal aa), were prepared,
and the phage titer (PFU per milliliter) was measured. ELISA plates were
coated with CH4-tail or CH4, respectively, and decreasing phage dilutions
were applied. Signals obtained from binding to CH4 were subtracted from
the CH4 tail signals as unspecific background. Shown are the longer HAX-1
cDNA clone (231 C-terminal aa) (f), the shorter cDNA clone (213 C-terminal
aa) (Œ), and the control phage, encoding an unrelated protein (E).
Titration curves of HAX-1-expressing phage in ELISA. We
FIGURE 5.
immunoprecipitation). A total of 2 ? 107J558L (murine myeloma cell line)
cells were lysed, and 30 ?g of the carrier-tail protein (CH4 tail) or carrier
protein (CH4) as negative control, respectively, were added to the lysates. The
His-tagged CH4 tail and CH4 were precipitated with anti-5xHis Ab. After
SDS-PAGE proteins were detected by immunoblotting. Upper panel, Devel-
oped with anti-5xHis Ab (CH4 tail and Ch4) and the lower panel with anti-
HAX-1 mAb (endogenous HAX-1). The lysates (lys) (in lanes 1 and 2) show
the carrier-tail protein (CH4 tail) or carrier protein (CH4), respectively, and the
endogenous HAX-1 expression. The precipitates (prec) are shown in lanes 3
and 4. HAX-1 coprecipitates with CH4 tail but not with CH4.
In vitro interaction of HAX-1 with the mIgE tail (in vitro co-
1142mIgE-INTERACTING PROTEINS

Page 5

added to the chilled and cleared cell lysates. As control, we per-
formed a precipitation with empty (CL-2B) Sepharose. Endoge-
nous HAX-1 coprecipitated with mIgE but could not be detected in
the control precipitation. As additional controls, we did the same
precipitations with JW813/4 BCR?cells. HAX-1 was not detect-
able in any of these control precipitations (Fig. 6).
siRNA-based knock down of HAX-1 in the B cell line JW813/4 BCR?
To elucidate a biological function of HAX-1 in the mIgE?B cell
line JW813/4 BCR?, we produced a HAX-1 knockdown by RNA
interference technology using siRNA. JW813/4 BCR?cells were
transfected with pU6-siRNA expression vectors for RNA interfer-
ence with HAX-1 expression. The two siRNA plasmids encoding
two different HAX-1 siRNA targets led to the expression of short
stem-loop RNAs homologous to 21 (target 1) or 19 (target 8) nu-
cleotides of the murine HAX-1 mRNA. The two HAX-1 targets
were transfected and after transfection single clones were isolated
by limiting dilution. Stable integration of the U6 promoter and the
siRNA target sequence was tested by PCR. Two PCR-positive
clones per target were further analyzed by real-time PCR (Fig. 7A)
and Western blot detection of HAX-1 (Fig. 7B), to test their ability
to interfere with HAX-1 expression on the mRNA and on the pro-
tein level. The clones transfected with target 1 were designated C1
and D1, and the clones transfected with target 8 were designated
A4 and D4. To monitor unspecific effects caused by siRNA-plas-
mid transfection, we additionally produced clones stably trans-
fected with a control target without homology to HAX-1, desig-
nated B1 and B3. Fig. 7A shows that the two clones transfected
with target 1 (C1 and D1) varied significantly in their HAX-1
expression. C1 reduced the HAX-1 mRNA level slightly but not
significantly compared with wild-type JW813/4 BCR?cells,
whereas D1 led to a 48% reduction in mRNA amounts. The two
clones transfected with target 8 (A4 and D4) led to a reduction of
mRNA of 90 and 93%, respectively, compared with the wild type.
The control target (clones B1 and B3) did not significantly influ-
ence HAX-1 mRNA expression.
HAX-1 protein expression in the siRNA transfectants was assessed
by Western blot detection. The protein expression levels (Fig. 7B)
align with the results of the real-time PCR measurements.
HAX-1 protein levels influence Ag-internalization efficiency
HAX-1 is associated with the cytoskeleton (15) and it was sug-
gested that HAX-1 participates in clathrin-mediated endocytosis of
the surface receptor BSEP in Madin-Darby canine kidney II cells
(22). Therefore, we asked whether the knockdown of HAX-1 ex-
pression in siRNA transfectants influenced the mIgE-mediated Ag
internalization in JW813/4 BCR?.
The JW813/4 BCR?transfectants with stable integrated HAX-
1-siRNA targets were tested for their ability to internalize BCR
after BCR stimulation with anti-IgE Ab. We found a correlation
between HAX-1 expression levels and the ability of the respective
clone to internalize the Ag. Clone C1 (target 1) showed no signif-
icant reduction of HAX-1 levels and in the internalization assay,
we measured the highest internalization efficiency of all tested
clones. Clone D1 (target 1) showed a 48% reduction of HAX-1
levels and in our BCR-internalization assay we measured approx-
imately half the values of clone C1. The two clones transfected
with target 8 (clones A4 and D4), showing a 90% reduction of
HAX-1 expression, exhibited the lowest efficiency in internaliza-
tion (Fig. 8). Table I demonstrates the correlation between HAX-1
levels and the internalization efficiency after 20 min. The internal-
ization of clone C1 was set to 100% and the reduction relative to
C1 was calculated for the other tested clones. For clone D1, a 48%
reduction in HAX-1 protein levels led to a reduction of 48% in
FIGURE 6.
coimmunoprecipitation). HAX-1 and the endogenous mIgE were copre-
cipitated form a murine B cell line expressing NP-specific mIgE (JW813/4
BCR?). The mIgE was precipitated with NP-conjugated Sepharose (NP).
Unconjugated Sepharose (CL-2B) was used as negative control. As second
negative control, we performed all precipitations also in the JW813/4
BCR?cell line, which does not express mIgE. Upper panel, Developed
with anti-murine IgE polyclonal rat serum and the lower panel with anti-
HAX-1 mAb. The lysates (lys) (in lanes 1 and 2) show the endogenous
HAX-1 expression in the JW813/4 cells with and without mIgE H chain,
designated as BCR?and BCR?, respectively. Lanes 3–6 show the pre-
cipitates (prec) with either NP-Sepharose or control (CL-2B)-Sepharose of
these two cell lines. Lane 8 in the upper panel shows the position of the IgE
band obtained with an IgE standard (arrow). HAX-1 coprecipitates with
mIgE and was not detected in any of the negative controls.
In vivo interaction of HAX-1 with the mIgE-tail (in vivo
FIGURE 7.
cells by RNA interference. A, Real-time PCR measurements of HAX-1
mRNA amounts in clones stably transfected with siRNA-expressing vec-
tors. HAX-1-mRNA levels were normalized using Arpb mRNA expression
levels. The fold change expresses the ratio of HAX-1 and Arpb mRNA
levels in each sample. “C1 and D1” represent clones transfected with
HAX-1-siRNA target 1; “A4 and D4” represent clones transfected with
HAX-1-siRNA target 8; “B1 and B3” represent clones transfected with the
control target; BCR?: wild-type JW813/4 expressing mIgE; BCR?:
JW813/4 without mIgE. C1 shows no significant reduction of HAX-1 ex-
pression compared with wild-type or control targets. D1 shows a reduction
of 48% and A4 and D4 show a 90 and 93% reduction. B, Immunoblots of
crude cell lysates of JW813/4 BCR?clones stably transfected with siRNA
expression vectors (pU6-siRNA). HAX-1-protein (lower band, arrow);
?-tubulin (upper band, arrow) was used as loading control. Immuoblots (B)
correlate with real-time PCR measurements (A).
“Knockdown” of HAX-1 expression in JW813/4 BCR?
1143The Journal of Immunology

Page 6

internalization efficiency. For target 8 (clones A4 and D4), this
correlation is not so explicit and the reduction of 90% in HAX-1
protein levels led to a 53% reduction in internalization efficiency.
Discussion
In the present study, we demonstrated in different independent ex-
periments the direct interaction of murine HAX-1 with the cyto-
plasmic tail of mIgE. Using phage display technology, we isolated
HAX-1-expressing phage clones and proved their specific binding
to the recombinant mIgE tail in direct ELISA experiments (Fig. 3).
Thus, explanations, which might favor a protein-protein-interac-
tion, supported or bridged by a third protein (e.g., reported for
Bcl-2-HAX-1-EBNA-LP (20)) could definitely be excluded. The
calculated KDof 7.4 ? 10?5M (Fig. 4), using SPRA, indicated a
biologically relevant association between HAX-1 and the cytoplas-
mic tail of mIgE. The biologically relevant interaction between
HAX-1 and the mIgE tail was further demonstrated with indepen-
dent coimmunoprecipitation experiments (Fig. 5: in vitro; Fig. 6:
in vivo). Because of our performed experimental setup, we con-
cluded that neither the in vitro nor the in vivo interaction between
HAX-1 and mIgE depends on any de novo protein modification
(i.e., phosphorylation). Finally, we measured the mIgE internal-
ization efficiency in mIgE?myeloma cells with reduced HAX-1
protein levels (Fig. 7) and clearly showed that reduction of HAX-
1-protein levels reduced the internalization efficiency (Fig. 8).
Already before, Tarlinton (31) anticipated that truncation of the
cytoplasmic tail of mIgG1 (32) and mIgE (33) would result in a
defect in Ag presentation. Indeed, the YANIL motif inside the
cytoplasmic tail of IgE (Fig. 9) matches the consensus sequence of
cytoplasmic internalization motifs and reflects the (single)-core
motif of an ITAM (YxxL/Ix6–8YxxL/I (34)). However, phosphor-
ylation of Y622in the cytoplasmic tail of IgE could never be
shown. Nevertheless, the Y622ANIL motif seems to play an im-
portant role in the interaction with HAX-1. Human HAX-1 (73%
homology and 69% identity with murine HAX-1) was shown to
interact with the K15 protein of Kaposi’s sarcoma-associated her-
pes virus via a conserved YASIL motif within the C terminus of
K15 (17). Due to the striking similarity between these two motifs,
it seems very likely that murine HAX-1 interacts with the murine
mIgE-tail via the YANIL-motif. Interestingly, as outlined in Fig. 9,
the YANIL motif is not present in the cytoplasmic tails of IgG iso-
types.ThismightindicatethattheinteractionbetweenHAX-1andthe
IgEcytoplasmictailisspecificforthisisotype.Thisassumption,how-
ever, needs detailed experimental approach in the future.
To define a first biological function of the mIgE-HAX-1 inter-
action, we focused on an experimental approach describing the
internalization capacity of mIgE with and without (or decreased)
HAX-1 expression. Indeed, Ortiz et al. (22) showed presence of
HAX-1 in clathrin-coated vesicles and HAX-1 participation in
clathrin-mediated endocytosis, which itself depends on the func-
tionality of cortactin. Cortactin is an F-actin-associated protein that
localizes within membrane ruffles in cultured cells, and is a direct
binding partner of the large GTPase dynamin. This direct interac-
tion suggests that cortactin may participate in one or several en-
docytic processes. Cao et al. (35) suggest that cortactin is an im-
portant component of the receptor-mediated endocytic machinery,
where, together with actin and dynamin, it regulates the scission of
clathrin pits from the plasma membrane. However, cortactin is not
expressed in normal B lymphocytes or plasma cells (36), but
Urono et al. (37) suggested that in hemopoietic cells, HS1 per-
forms a cortactin-homologous function by modulating the actin
assembly. The structure of HS1 is closely related to that of cor-
tactin although HS1 contains only 3.5 repeat units, in contrast to
5.5 or 6.5 repeat units found in cortactin isoforms. Urono et al. (37)
showed that HS1 is able to bind to F-actin with modest affinity and
furthermore, that HS1 is able to interact directly with actin-related
protein (Arp)2/3 complex and activate Arp2/3 complex-mediated ac-
tin nucleation and branching in a manner similar to cortactin. Thus it
seems very likely that hemopoietic cells encompass a unique actin-
polymerization system that involves HS1 and the Arp2/3 complex.
Interestingly, HAX-1 is a binding partner of both, cortactin and
HS1, which explains our interest in the influence of HAX-1 on
receptor-mediated internalization of Ag. Indeed, HAX-1-depletion
retarded BCR internalization (Fig. 8). A 48% reduction of HAX-1
level led to a 48% reduction of internalization efficiency and a 90%
reduction of HAX-1 levels reduced the internalization efficiency
by 53%. From our point of view, the reduced internalization effi-
ciency in answer to HAX-1 reduction explains the phenotypes of
FIGURE 8.
(JW813/4 BCR?) stably transfected with HAX-1 siRNA targets. IgE-BCR
was stimulated with anti-IgE-FITC Ab (EM95-3-FITC) for the indicated
time. FITC decreasing fluorescence is due to the pH shift upon internal-
ization. The decrease of fluorescence over time was measured. One hun-
dred percent fluorescence is the fluorescence at time “0” meaning 0% in-
ternalization. No fluorescence is calculated as 100% internalization. The
reduction of HAX-1 protein levels led to a reduced efficiency of Ag inter-
nalization. Values were calculated from ten independent experiments.
HAX-1-siRNA target 1 clone C1 (?), HAX-1-siRNA target 1 clone D1
(F) and HAX-1-siRNA target 8 (clones A4 and D4) (Œ) as indicated.
BCRinternalization in mIgE-expressingcelllines
Table I. Ag internalization after 20-min incubation correlates with
HAX-1 mRNA levels
Target/Clone
% HAX-1
Expressiona
Internalization
(%)b
% Internalization
Relative to C1c
1/C1
1/D1
8/A4 and D4
95–100
52
10
51.4
27.0
24.2
100
52.4
47.0
aHAX-1 expression was calculated as the percentage of mRNA levels from wild-
type cells.
bThe internalization (in percent) was calculated based on the decrease of fluo-
rescence for each HAX-1-siRNA target clone.
cClone C1 shows unchanged HAX-1 levels compared to wild-type cells or control
siRNA-target clones. Therefore, the internalization frequency of this clone was 100%,
and we calculated the relative internalization values from the other HAX-1-siRNA-
transfected clones (D1, A4, and D4) accordingly.
FIGURE 9.
isotypes. Amino acid residues, not matching the IgE sequence are boxed.
The YANIL motif is marked in gray.
Sequence alignment of the cytoplasmic tails of murine Ig
1144mIgE-INTERACTING PROTEINS

Page 7

our knockout mouse strain IgEKVK?tail, expressing a truncated cy-
toplasmic tail of three amino acids (K-V-K). The truncation of the
tail resulted in a 50% decrease of serum IgE, a weak secondary
response, linked with a missing affinity maturation for IgE Abs
(38). At that time, we concluded that the reduced IgE titers solely
reflect the loss of biological activities associated with the cyto-
plasmic domain of IgE. Two hypotheses could be brought forward.
First, signals generated via mIg are needed at all times, not only for
the maturation process but also for the expansion of Ag-specific
cells. Second, Ag presentation to Th cells is necessary during an
Ab response, and only the Ag receptor is capable of effective Ag
capture for presentation (33, 38). In the present study, we suggest that
the incapability of the truncated tail to bind HAX-1 and therewith
linked decreased Ag receptor internalization efficiency is responsible
for the observed decreased serum IgE levels in IgEKVK?tail.
Summarizing, we suggest a model in which HAX-1 physically
links the mIgE molecule to the cytoskeleton via its interaction with
HS1. This link would be crucial for the efficient internalization of
Ag in mIgE?B cells. In B cells with a truncated mIgE tail (33),
BCR internalization would be less efficient due to the absence of
this link resulting in decreased Ag presentation to T cells. Espe-
cially in germinal centers, where B cells undergo rapid prolifera-
tion, it might be necessary to achieve rapid peptide loading to get
T cell help in every cell cycle. This model is also in agreement
with the fact that tail-knockout mice (33) show defects in the se-
lection of high-affinity Abs in the hypermutation process in ger-
minal centers (38). In these knockout mice, even hypermutated B
cells that did retain the specificity for the Ag carrying the respec-
tive T cell epitope would not get sufficient specific T cells support
for repeated rounds of positive selection.
From the theoretical point of view, cytoplasmic tails of mIgs
might commit signals transmitted by 1) the tail alone, by 2) steri-
cally influencing the binding of adaptor proteins to the Ig?-Ig?
coat proteins, or by 3) their capacity to actively bind proteins. With
our current report, we clearly supported the hypothesis for the
existence of actively cytoplasmic tail-binding proteins.
Disclosures
The authors have no financial conflict of interest.
References
1. Venkitaraman, A. R., G. T. Williams, P. Dariavach, and M. S. Neuberger. 1991.
The B-cell antigen receptor of the five immunoglobulin classes. Nature 352:
777–781.
2. Reth, M. 1992. Antigen receptors on B lymphocytes. Annu. Rev. Immunol. 10:
97–121.
3. Reth, M., and J. Wienands. 1997. Initiation and processing of signals from the B
cell antigen receptor. Annu. Rev. Immunol. 15: 453–479.
4. Ravetch, J. V., and L. L. Lanier. 2000. Immune inhibitory receptors. Science 290:
84–89.
5. Geisberger, R., G. Achatz-Straussberger, I. Oberndorfer, D. Infuhr, R. Crameri,
and G. Achatz. 2004. Isolation of a serine threonine kinase interacting with the
cytoplasmic tail of mIgE. Allergy Clin. Immunol. Int. 1: 177–181.
6. Geisberger, R., R. Crameri, and G. Achatz. 2003. Models of signal transduction
through the B-cell antigen receptor. Immunology 110: 401–410.
7. Geisberger, R., M. Prlic, G. Achatz-Straussberger, I. Oberndorfer, E. Luger,
M. Lamers, R. Crameri, U. Appenzeller, J. Wienands, M. Breitenbach, et al.
2002. Phage display based cloning of proteins interacting with the cytoplasmic
tail of membrane immunoglobulins. Dev. Immunol. 9: 127–134.
8. Engels, N., B. Wollscheid, and J. Wienands. 2001. Association of SLP-65/BLNK
with the B cell antigen receptor through a non-ITAM tyrosine of Ig-?. Eur.
J. Immunol. 31: 2126–2134.
9. DeFranco, A. L. 1997. The complexity of signaling pathways activated by the
BCR. Curr. Opin. Immunol. 9: 296–308.
10. Kurosaki, T. 1999. Genetic analysis of B cell antigen receptor signaling. Annu.
Rev. Immunol. 17: 555–592.
11. Yamanashi, Y., T. Fukuda, H. Nishizumi, T. Inazu, K. Higashi, D. Kitamura,
T. Ishida, H. Yamamura, T. Watanabe, and T. Yamamoto. 1997. Role of tyrosine
phosphorylation of HS1 in B cell antigen receptor-mediated apoptosis. J. Exp.
Med. 185: 1387–1392.
12. Yamanashi, Y., M. Okada, T. Semba, T. Yamori, H. Umemori, S. Tsunasawa,
K. Toyoshima, D. Kitamura, T. Watanabe, and T. Yamamoto. 1993. Identifica-
tion of HS1 protein as a major substrate of protein-tyrosine kinase(s) upon B-cell
antigen receptor-mediated signaling. Proc. Natl. Acad. Sci. USA 90: 3631–3635.
13. Suzuki, Y., C. Demoliere, D. Kitamura, H. Takeshita, U. Deuschle, and
T. Watanabe. 1997. HAX-1, a novel intracellular protein, localized on mitochon-
dria, directly associates with HS1, a substrate of Src family tyrosine kinases.
J. Immunol. 158: 2736–2744.
14. Cilenti, L., M. M. Soundarapandian, G. A. Kyriazis, V. Stratico, S. Singh, S. Gupta,
J. V. Bonventre, E. S. Alnemri, and A. S. Zervos. 2004. Regulation of HAX-1 anti-
apoptotic protein by Omi/HtrA2 protease during cell death. J. Biol. Chem.
15. Gallagher, A. R., A. Cedzich, N. Gretz, S. Somlo, and R. Witzgall. 2000. The
polycystic kidney disease protein PKD2 interacts with Hax-1, a protein associ-
ated with the actin cytoskeleton. Proc. Natl. Acad. Sci. USA 97: 4017–4022.
16. Rogers, S., R. Wells, and M. Rechsteiner. 1986. Amino acid sequences common
to rapidly degraded proteins: the PEST hypothesis. Science 234: 364–368.
17. Sharp, T. V., H. W. Wang, A. Koumi, D. Hollyman, Y. Endo, H. Ye, M. Q. Du,
and C. Boshoff. 2002. K15 protein of Kaposi’s sarcoma-associated herpesvirus is
latently expressed and binds to HAX-1, a protein with antiapoptotic function.
J. Virol. 76: 802–816.
18. Kawaguchi, Y., K. Nakajima, M. Igarashi, T. Morita, M. Tanaka, M. Suzuki,
A. Yokoyama, G. Matsuda, K. Kato, M. Kanamori, and K. Hirai. 2000. Interac-
tion of Epstein-Barr virus nuclear antigen leader protein (EBNA-LP) with HS1-
associated protein X-1: implication of cytoplasmic function of EBNA-LP. J. Vi-
rol. 74: 10104–10111.
19. Dufva, M., M. Olsson, and L. Rymo. 2001. Epstein-Barr virus nuclear antigen 5
interacts with HAX-1, a possible component of the B-cell receptor signalling
pathway. J. Gen. Virol. 82: 1581–1587.
20. Matsuda, G., K. Nakajima, Y. Kawaguchi, Y. Yamanashi, and K. Hirai. 2003.
Epstein-Barr virus (EBV) nuclear antigen leader protein (EBNA-LP) forms com-
plexes with a cellular anti-apoptosis protein Bcl-2 or its EBV counterpart BHRF1
through HS1-associated protein X-1. Microbiol. Immunol. 47: 91–99.
21. Yedavalli, V. S., H. M. Shih, Y. P. Chiang, C. Y. Lu, L. Y. Chang, M. Y. Chen,
C. Y. Chuang, A. I. Dayton, K. T. Jeang, and L. M. Huang. 2005. Human im-
munodeficiency virus type 1 Vpr interacts with antiapoptotic mitochondrial pro-
tein HAX-1. J. Virol. 79: 13735–13746.
22. Ortiz, D. F., J. Moseley, G. Calderon, A. L. Swift, S. Li, and I. M. Arias. 2004.
Identification of HAX-1 as a protein that binds bile salt export protein and reg-
ulates its abundance in the apical membrane of Madin-Darby canine kidney cells.
J. Biol. Chem. 279: 32761–32770.
23. Radhika, V., D. Onesime, J. H. Ha, and D. N. Dhanasekaran. 2004. G? 13
stimulates cell migration through cortactin-interacting protein Hax-1. J. Biol.
Chem. 279: 49406–49413.
24. Al-Maghrebi, M., H. Brule, M. Padkina, C. Allen, W. M. Holmes, and
Z. E. Zehner. 2002. The 3? untranslated region of human vimentin mRNA inter-
acts with protein complexes containing eEF-1? and HAX-1. Nucleic Acids Res.
30: 5017–5028.
25. Yin, H., H. Morioka, C. A. Towle, M. Vidal, T. Watanabe, and L. Weissbach.
2001. Evidence that HAX-1 is an interleukin-1? N-terminal binding protein.
Cytokine 15: 122–137.
26. Schwarzbaum, S., A. Nissim, I. Alkalay, M. C. Ghozi, D. G. Schindler,
Y. Bergman, and Z. Eshhar. 1989. Mapping of murine IgE epitopes involved in
IgE-Fc? receptor interactions. Eur. J. Immunol. 19: 1015–1023.
27. Sheffield, P., S. Garrard, and Z. Derewenda. 1999. Overcoming expression and
purification problems of RhoGDI using a family of “parallel” expression vectors.
Protein Expr. Purif. 15: 34–39.
28. Crameri, R., G. Achatz, M. Weichel, and C. Rhyner. 2002. Direct selection of
cDNAs by phage display. Methods Mol. Biol. 185: 461–469.
29. Crameri, R., and M. Suter. 1993. Display of biologically active proteins on the
surface of filamentous phages: a cDNA cloning system for selection of functional
gene products linked to the genetic information responsible for their production.
Gene 137: 69–75.
30. Aluvihare, V. R., A. A. Khamlichi, G. T. Williams, L. Adorini, and
M. S. Neuberger. 1997. Acceleration of intracellular targeting of antigen by the
B-cell antigen receptor: importance depends on the nature of the antigen-antibody
interaction. EMBO J. 16: 3553–3562.
31. Tarlinton, D. 1997. Antigen presentation by memory B cells: the sting is in the
tail. Science 276: 374–375.
32. Kaisho, T., F. Schwenk, and K. Rajewsky. 1997. The roles of ?1 heavy chain mem-
brane expression and cytoplasmic tail in IgG1 responses. Science 276: 412–415.
33. Achatz, G., L. Nitschke, and M. C. Lamers. 1997. Effect of transmembrane and
cytoplasmic domains of IgE on the IgE response. Science 276: 409–411.
34. Reth, M. 1989. Antigen receptor tail clue. Nature 338: 383–384.
35. Cao, H., J. D. Orth, J. Chen, S. G. Weller, J. E. Heuser, and M. A. McNiven.
2003. Cortactin is a component of clathrin-coated pits and participates in recep-
tor-mediated endocytosis. Mol. Cell. Biol. 23: 2162–2170.
36. Miglarese, M. R., J. Mannion-Henderson, H. Wu, J. T. Parsons, and T. P. Bender.
1994. The protein tyrosine kinase substrate cortactin is differentially expressed in
murine B lymphoid tumors. Oncogene 9: 1989–1997.
37. Uruno, T., P. Zhang, J. Liu, J. J. Hao, and X. Zhan. 2003. Haematopoietic lineage
cell-specific protein 1 (HS1) promotes actin-related protein (Arp) 2/3 complex-
mediated actin polymerization. Biochem. J. 371: 485–493.
38. Luger, E., M. Lamers, G. Achatz-Straussberger, R. Geisberger, D. Infuhr,
M. Breitenbach, R. Crameri, and G. Achatz. 2001. Somatic diversity of the im-
munoglobulin repertoire is controlled in an isotype-specific manner. Eur. J. Im-
munol. 31: 2319–2330.
1145 The Journal of Immunology